Wyniki wyszukiwania - BazTech

1

Lexicon and attention based handwritten text recognition system

Kumari Lalita, Singh Sukhdeep, Rathore Vaibhav Varish Singh, Sharma Anuj

Machine Graphics and Vision

|

2022

|

Vol. 31, No. 1/4

75--92

EN

The handwritten text recognition problem is widely studied by the researchers of computer vision community due to its scope of improvement and applicability to daily lives. It is a sub-domain of pattern recognition. Due to advancement of computational power of computers since last few decades neural networks based systems heavily contributed towards providing the state-of-the-art handwritten text recognizers. In the same direction, we have taken two state-of-the art neural networks systems and merged the attention mechanism with it. The attention technique has been widely used in the domain of neural machine translations and automatic speech recognition and now is being implemented in text recognition domain. In this study, we are able to achieve 4.15% character error rate and 9.72% word error rate on IAM dataset, 7.07% character error rate and 16.14% word error rate on GW dataset after merging the attention and word beam search decoder with existing Flor et al. architecture. To analyse further, we have also used system similar to Shi et al. neural network system with greedy decoder and observed 23.27% improvement in character error rate from the base model.

2

Signature analysis system using a convolutional neural network

Winnicka Alicja, Kęsik Karolina, Połap Dawid

Annals of Computer Science and Information Systems

|

2019

|

Vol. 18

287--290

EN

Identity verification using biometric methods has been used for many years. A special case is a handwritten signature made on a digital device or piece of paper. For the digital analysis and verification of its authenticity, special methods are needed. Unfortunately, this is a rather complicated task that quite often requires complex processing techniques. In this paper, we propose a system of signatures verification consisting of two stages. In the first one, a signature pattern is created. Thanks to this, the first attempt to verify identity takes place. In the case of approval, the second stage is followed by the processing of a graphic sample containing a signature by the convolutional neural network. The proposed technique has been described, tested and discussed due to its practical use.

3

Metody ekstrakcji linii tekstu w złożonych dokumentach rękopiśmiennych

Pach J.

Elektronika : konstrukcje, technologie, zastosowania

|

2015

|

Vol. 56, nr 12

92-96

PL

W artykule przedstawiono metody komputerowe wykorzystywane do wykrywania linii tekstu w dokumentach rękopiśmiennych. Przedstawiono problematykę automatycznej identyfikacji autora tekstu na podstawie cech jego pisma. Ponieważ jest to problematyka złożona, omówiono ogólną metodologię przetwarzania tekstu z wykorzystaniem przetwarzania cyfrowej wersji obrazu dokumentu zeskanowanego lub pozyskanego poprzez fotografię. Omówiono główne grupy algorytmów służących do wykrywania linii w tekście, przedstawiając ich ogólną ideę, wady i zalety. Zaprezentowano także autorski algorytm wykorzystujący transformatę Hougha, którego skuteczność analizy trudnych średniowiecznych dokumentów łacińskich jest wyższa, niż pozostałych podejść. Wykazano jej dokładność na przykładzie eksperymentu z wybranymi dokumentami archiwalnymi.

EN

The paper presents the computer-based methods for the text line detection in hand-written manuscripts. The problem of the automated author detection based on his writing habits was defined. Because the task is difficult and complex, the general text processing methodology is introduced, working with the scanned or photographed documents. The main groups of algorithms applied to the text line detection were introduced with their advantages and drawbacks iterated. The novel approach for the task, exploiting the modified Hough transform is also presented. Its efficiency of detecting text lines in the complex medieval manuscripts is higher than for approaches used so far. This is demonstrated based on the selected archived documents.

4

Handwriting recognition in intelligent design systems

Kacalak W., Majewski M.

Archives of Mechanical Technology and Automation

|

2014

|

Vol. 34, no. 3

3--12

EN

This article presents selected research on the development of complex fundamentals of building intelligent interactive systems for design of machine elements and assemblies on the basis of its features described in a natural language. We propose a new method for handwriting recognition that utilizes geometric features of letters. The article deals with recognition of isolated handwritten characters using neural networks. As a result of the geometrical analysis, graphical representations of recognized characters are obtained in the form of pattern descriptions of isolated characters. Selected parameters of the characters are inputs to the neural network for writing recognition which is font independent. In this article, we present a new method for off-line natural writing recognition and also describe our research and conclusions on the experiments.

PL

W artykule przedstawiono wybrane prace badawcze dotyczące podstaw budowy inteligentnych systemów interakcji do projektowania elementów i zespołów maszyn na podstawie ich cech opisywanych w języku naturalnym. Zaproponowano nową metodę rozpoznawania pisma odręcznego, w której wykorzystano geometryczne cechy znaków. Artykuł dotyczy rozpoznawania izolowanych znaków pisma odręcznego za pomocą sieci neuronowych. W wyniku analizy geometrycznej otrzymuje się reprezentacje graficzne rozpoznawanych znaków w postaci opisów wzorców pojedynczych znaków. Wybrane parametry znaków stanowią wejścia sieci neuronowej do rozpoznawania pisma niezależnego od kroju. W artykule przedstawiono nową metodę rozpoznawania pisma naturalnego, a także opisano badania i podano wnioski wynikające z eksperymentów.

5

Wybrane problemy efektywnego rozpoznawania pisma odręcznego

Kacalak W., Majewski M.

Pomiary Automatyka Kontrola

|

2011

|

R. 57, nr 5

479-482

PL

W artykule przedstawiono nowatorską metodę efektywnego rozpoznawania pisma odręcznego z zastosowaniem opracowanych sposobów analiz geometrycznych znaków i wybranych metod sztucznej inteligencji. Proponowana metoda analiz geometrycznych znaków oparta na opracowanym sposobie odpowiednich pomiarów odległości wybranych ich punktów pozwala na rozpoznawanie pisma odręcznego niezależnie od stylu i charakteru pisma operatora. W rezultacie zastosowania metody otrzymuje się zakodowaną reprezentację znaku dla efektywnego rozpoznawania przez sztuczne sieci neuronowe. Artykuł również przedstawia system rozpoznawania odręcznego pisma operatora zbudowany z podsystemów wstępnego przetwarzania, analiz geometrycznych, logiki rozmytej, sieci neuronowych oraz ich wyspecjalizowanych modułów. Proponowany inteligentny system może stanowić nowoczesny i efektywny system interakcji urządzeń technicznych i ich operatorów w zadaniach sterowania.

EN

In this paper, an innovative method for effective handwriting recognition is presented. It uses the developed methods of geometrical analyses of isolated handwritten characters and selected artificial intelligence methods. The proposed geometrical feature analysis method, based on the developed manner of appropriate measurements of distances of selected character points, allows handwriting recognition independent of different writing and character styles, and writing conditions. As a result of using the method, encoded representations of characters are obtained for effective recognition by artificial neural networks. The paper also presents an operator's handwriting recognition system consisting of the subsystems of preprocessing, geometrical analyses, fuzzy logic, neural networks, and their specialized modules. Handwriting recognition has always been a challenging problem for artificial intelligence researchers, and remains an open issue. It is because of the complexity of the handwriting recognition task. The intelligent handwriting recognition system of the technical device operator's natural writing can be a modern and effective interaction system [3, 5]. In the paper, a review of selected issues is carried out with regards to the handwriting recognition issues, new geometrical analysis method (fig. 1, 2, 3, 4) and concept of a handwriting recognition system (fig. 5, 6). The proposed system is novel in that it integrates efficient geometrical processing with artificial intelligence methods to use neural networks and fuzzy logic for effective handwriting recogni-tion.

6

HMM-based Online Handwritten Gurmukhi Character Recognition

Sharma A., Kumar R., Sharma R. K.

Machine Graphics and Vision

|

2010

|

Vol. 19, No. 4

439-449

EN

This paper presents a hidden Markov model-based online handwritten character recognition for Gurmukhi script. We discuss a procedure to develop a hidden Markov model database in order to recognize Gurmukhi characters. A test with 60 handwritten samples, where each sample includes 41 Gurmukhi characters, shows a 91.95% recognition rate, and an average recognition speed of 0.112 seconds per stroke. The hidden Markov model database has been developed in XML using 5330 Gurmukhi characters. This work shall be useful to implement a hidden Markov model in online handwriting recognition and its software development.

7

Multistage semi-automatic text image segmentation for training set acquisition in handwriting recognition

Sas J.

Systems Science

|

2008

|

Vol. 34, no 1

107-126

EN

In the paper, a complete method of text image segmentation into the images of individual characters is proposed. The ultimate aim of the segmentation process is to prepare a set of correctly labeled character samples that can be used to train the character classifier applied as the component of the handwritten word recognizer. The method proposed consists of two stages. At the first stage, the text image is first divided into lines and then the lines are segmented into words. In this phase, the known spelling representation of the text on the image is used, so as to obtain as many segments as the number of words in the text. The information about the expected width of known words is also utilized. At the second stage, the obtained images of known words are segmented into individual characters. The multiphase procedure is applied. It first segments individual words independently, using the estimates of character widths obtained by the complete text corpus analysis. Then the global text segmentation is elaborated, which maximizes the similarity measures of samples extracted for all alphabet characters. Genetic algorithm is applied in this phase. Finally, the segmentation variants represented by chromosomes in the terminal population of the genetic algorithm are locally refined and the most dissimilar samples in sets corresponding to the alphabet characters are rejected. The experiments conducted showed that the accuracy of handwriting recognition achieved by recognizers trained with the training set obtained with the proposed method is close to the accuracy achievable with the training set prepared by a human expert.

8

O Ukrytych Modelach Markowa i ich zastosowaniu do rozpoznawania ciągów znaków pisma odręcznego

Klęsk P.

Metody Informatyki Stosowanej

|

2008

|

nr 1 (Tom 13)

175-189

EN

In the paper, selected informations on Hidden Markov Models (also called Hidden Markov Chains) are reminded. Basic riotions are defined and algorithms related to these models are shortly presented. The research part of the papers shows results of three conducted experiments entitled: "pork cutlet", "form sheet" and ''poetaster". The most important experiment "form sheet" gives a good starting point to a practical application of HMMs to the. handwriting recognition. The "poetaster" experiment shows possible application of HMMs in so called "artifial creation".

9

Hidden Markov Model (HMM) - opis modelu i algorytmów pod kątem wykorzystania w problemach rozpoznawania mowy i pisma

Gąciarz T.

Elektrotechnika i Elektronika

|

1998

|

T. 17, nr 1

17--31

EN

The use of hidden Markov models (HMM's) for speech and handwriting recognition has become increasingly popular in the past few years. The reason why this method has become so popular are: the inherent statistical (mathematically precise) framework, the easy and availability of training algorithms for estimating the parameters of the models from the finite training sets of data, the flexibility of the resulting recognition system where one can easily change the size, type, or architecture of the models to suit particular words, sounds etc., and the ease of implementation of the overall recognition system. In this paper, the basic information on hidden Markoy model is presented. It includes formal description of the model, methods and algorithms used for training and recognition.

PL

Hidden Markov Model już od kilkunastu lat cieszy się niesłabnącą popularnością w zastosowaniach związanych z rozpoznawaniem mowy i pisma ręcznego. Przyczyną tak wielkiej popularności są niewątpliwie solidnie opracowane podstawy matematyczne modelu (teoria), jak i struktura implementacyjna (praktyka). Wykorzystywane, ogólnie dostępne algorytmy charakteryzują się jasnością i efektywnością estymacji parametrów modelu na podstawie skończonego zbioru uczącego. Elastyczność systemu rozpoznawania pozwala na łatwe dostosowanie rozmiaru, typu i architektury modelu do odpowiednich słów, dźwięków itp. W artykule tym zawarto podstawowe dane na temat budowy HMM wraz z jego formalnym opisem. Przedstawiono także ideę działania najbardziej znanych algorytmów służących do uczenia i rozpoznawania z wykorzystaniem modelu.